Cluster, Grid, Cloud Concepts Kalaiselvan.K
Contents Section 1: Cluster Section 2: Grid Section 3: Cloud
Cluster An Overview Need for a Cluster Cluster categorizations
A computer cluster is a group of linked computers, working together closely thus in many respects forming a single computer. The components of a cluster are connected to each other through fast local area networks What is a Cluster?
Need for a Cluster Requirements for computing increasing fast. More data to process. More compute intensive algorithms available. Approaches to supply demand Qualitative: Optimized algorithms, faster processors, more memory. Quantitative: Cluster computing, grid computing, etc.
Cluster categorizations High Availability Cluster Load Balancing Cluster HPC Cluster
High Availability Clusters Failover Clusters, mainly implemented to improve the availability of service that cluster provides They operates by having redundant nodes, upon failure the standby node take cares Types of High availability clusters: one way & two way Often used for critical databases, network file sharing and business applications
Load Balancing Clusters Multiple computers connected together to share computational workload Logically they are multiple computers but function as single virtual computer Request initiated from the user is distributed among all the nodes by one or more load balancer
HPC Clusters HPC clusters are mainly used to increases the performance by splitting the computational task into different nodes Mainly used in scientific computing Popular HPC cluster implementations are nodes running with linux os and free software s to implement the parallelism
Contd. The job running on the cluster nodes requires little or no inter nodes communication is called Grid Computing The local Scheduling software manages the cluster nodes load balancing Middleware such as MPI (Message Passing Interface) or PVM (Parallel Virtual Machine) permits compute clustering programs to be portable to a wide variety of clusters
Grid What is Grid? Why Grid? Grid Architecture Virtual Organization Grid Middleware's & It s Functionalities Grid Applications
What is Grid? Definition: Grid computing is a term referring to the combination of computer resources from multiple administrative domains to reach a common goal. Coordinates resources that are not subject to centralized control Uses standard, open, general-purpose protocols and interfaces Delivers nontrivial qualities of service
Why Grid? Large-scale science and engineering are done through the interaction of people, heterogeneous computing resources, information systems, and instruments, all of which are geographically and organizationally dispersed. The overall motivation for Grids is to facilitate the routine interactions of these resources in order to support large-scale science and Engineering.
Virtual Organization Virtual Organization (VO) refers to a dynamic set of individual and/or institutions defined around a set of resource-sharing rules and conditions Multiple organizations that function as one unit through the use of their shared competencies and resources for the purpose of one or more identified goals Example: LHC :1800 Physicists, 150 Institutes, 32 Countries 100 PB of data by 2010; 50,000 CPUs
Components of Grid Nagios, Monalisa Ganglia,GridICE Bioinformatics,Disaster management Applications Grid Portals, PSE s, API s Access Methods Debuggers, Compiler IDE, Profilers, Workflow tools SRB/SRM, Visualization, softwares Globus/ Glite/ Unicore, Torque/Loadleveler/SGE, Gridway/MOAB/Condor-G, GSI, Grid services, MPICH & MPICH-G2 Program Development Environment Storage & Visualization Scheduler, Middleware & Security Monitoring & Management CPU, Satellite terminals, Telescopes Computational Resources & devices GARUDA, NKN and Internet links Network/Communication Fabric
Grid Architecture Grid Architecture can be described as the layers of building blocks, where each layer has a specific function, to accomplish Grid Computing Infrastructure
Grid Architecture
Grid Middleware A mediator layer that provide a consistent and homogeneous access to resources managed locally with different syntax and access method It provide an uniform interface of the Grid to users and handle all the complexity generated due to heterogeneous systems. Middleware S/W is a layer between grid applications and low level functionality of grid
Popular Grid Middleware Globus Toolkit Globus Alliance Glite-EGEE Gridbus-University of Melbourne Unicore (Uniform Interface to Computing Resources )-Institute for Advanced Simulation, Guelich, Germany OMII from the Open Middleware Infrastructure Institute
Middleware functionalities Security Job Management Data Management Information Management
Security Information Security Secure communication Authentication Single sign on & Delegation Authorization Resource Level VO Level Infrastructure Level Security Host Security
Job Management Support an open Job Description Language RSL, JDL, JSDL Submission, Status Query, Cancel & Destroy, Getting Output & Error Transferring input/output data from/to remote source/destination Support Serial/ Parallel Jobs (Heterogeneous & Homogeneous) Integration with all Local Resource Managers
Grid Middleware: Data Management Two Basic Categories of Data Management Data Movement Secure Robust Efficient Third party movement Data Replication One or more copies or replicas Survive loss Easy availability Reduce access latency Performance for distributed applications
Grid Information Management System information is critical to operation of the grid and construction of applications How does an application determine what resources are available? What is the state of the computational grid? How can we optimize an application based on configuration of the underlying system? We need a general information infrastructure to answer these questions
Grid Information System Provides mechanism for discovery and monitoring of resources Designed to provide various characteristics of resource, computation, service and other entities. Provide access to static and dynamic information regarding system components Access to information is subject to authentication and authorization mechanisms. Information sources are distributed
Grid Applications Types : Sequential Jobs for particular platform Concurrent Sequential Jobs for different platforms Homogeneous Parallel job for particular OS Heterogeneous Parallel Jobs
Grid Applications Bio Informatics applications High Energy Physics Applications Weather Modelling and Predicting Ocean Currents Disaster Management Aerodynamic Simulations
Advantages/ Disadvantages Advantages Can solve larger, more complex problems in a shorter time Easier to collaborate with other organizations Make better use of existing hardware Disadvantages Grid software and standards are still evolving Learning curve to get started Non-interactive job submission
Cloud Computing Cloud Overview What is New in Cloud? Cloud Architecture Cloud Types Cloud Services Cloud Virtualization Challenges
Cloud Overview Definition: A large-scale distributed computing paradigm that is driven by economies of scale, in which a pool of abstracted, virtualized, dynamically-scalable, managed computing power, storage, platforms, and services are delivered on demand to external customers over the Internet.
Contd.. Massively scalable Can be encapsulated as an abstract entity that delivers different levels of service Driven by economies of scale Services can be dynamically configured (via virtualization or other approaches) and delivered on demand increase in computing power and storage capacity (multicores etc) Exponentially growing data size Widespread adoption of Services Computing and Web 2.0 apps
What is New in Cloud Computing? It is sold on demand minute or a hour It is Elastic can add or remove the resources at any time Fully Managed by the Cloud computing provider Infinite availability of resources
Cloud Computing Types Public Clouds Resources are dynamically provisioned on fine grained and self service basis over internet via web application & web pages to the public Standard computing model The services are dynamically scalable and often billable
Cloud Computing Types Private Cloud Cloud computing implemented within corporate firewall Offers same features of public clouds Free from control over data, security issues and regulatory compliances
Cloud Computing Types Hybrid Clouds Organization provides and manages some resources in house and other s are from public cloud. Benefits: Organizations who are already having infrastructure Who want keep their sensitive data on their own control
General Cloud Architecture
Infrastructure as a Service Infrastructure as a Service (IaaS): hw, sw, equipments, can scale up and down dynamicallly (elastic). E.g.: Amazon Elastic Compute Cloud (EC2) and Simple Storage Service (S3) Eucalyptus: open source Cloud implementation compatible with EC2 (allows to set up local cloud infra prior to buying services)
Platform as a Service Platform as a Service (PaaS): offers high level integrated environment to build, test, and deploy custom apps. Restrictions on sw used to develop apps in exchange for built-in scalability. E.g.: Google App Engine
Software as a Service Software as a Service (SaaS): delivers special purpose software that is remotely accessible. E.g,: Google Maps, Live Mesh from Microsoft etc
Virtualization
Cloud Virtualization Virtualization is a virtualization of computers or OS Hides the physical characteristics of a computing platform Multi Tenancy Multiple users can share the same physical resource Lower no of Physical Servers server consolidation space utilization Multiple OS can run on single machine
Challenges Data Residency Data Location Data Segregation Security Interoperability between different cloud
Thank You
Secure communication : GSI uses Public key cryptography message is encrypted using public key. And decrypted using private key Authentication : user or service are authenticated using grid certificate Single sign on : reduces no of time sign in, done using proxy, done by certificate holder, it has time notation Authrization VO Level auth system : RP will have the rights, user gets credintial from authrization system, and send it to the RP to accquire resource
Similarities & Difference Reduced Cost: The price of off-the-shelf consumer desktops has plummeted in recent years, and thi The average desktop PC today is many times mor 2. Processing Power : The parallel processing power of a high-performance cluster c This reduced price per unit of power enables enterp Improved Network Technology: Driving the development of computer clust along with a reduction in the clusters are typically connected via a single virtual local area network (VLAN), and the network treats each computer as a separa ility: Perhaps the greatest advantage of computer clusters is the scalability they offer. While mainframe computers have a fixed 5. Availability: When a mainframe computer fails, the entire system fails. However, if a node in a computer cluster fails, its